Published August 15, 1996
unc-68 Encodes a Ryanodine Receptor Involved in Regulating C. elegans
Body-Wall Muscle Contraction
Ed B. Maryon, Roberto Coronado,* and Philip Anderson
Department of Genetics, and Department of Physiology, *University of Wisconsin, Madison, Wisconsin 53706
Abstract. Striated muscle contraction is elicited by the
release of stored calcium ions through ryanodine receptor channels in the sarcoplasmic reticulum, ryr-1 is a
C. elegans ryanodine receptor homologue that is expressed in body-wall muscle cells used for locomotion.
Using genetic methods, we show that ryr-1 is the previously identified locus uric-68. First, transposon-induced
deletions within ryr-1 are alleles of unc-68. Second,
transformation of unc-68 mutants with ryr-1 genomic
D N A results in rescue of the Unc phenotype, unc-68
mutants move poorly, exhibiting an incomplete flaccid
paralysis, yet have normal muscle ultrastructure. The
mutants are insensitive to the paralytic effects of ryanodine, and lack detectable ryanodine-binding activity.
The Unc-68 phenotype suggests that ryanodine receptors are not essential for excitation-contraction coupling in nematodes, but act to amplify a (calcium) signal
that is sufficient for contraction.
© The Rockefeller University Press, 0021-9525/96/08/885/9 $2.00
The Journal of Cell Biology, Volume 134, Number 4, August 1996 885-893
885
I
Please address all correspondence to P. Anderson, Department of Genetics, University of Wisconsin, Madison, Wisconsin 53706. Tel.: (608) 2638429, Fax: (608) 262-2976. E-mail: [email protected]
Downloaded from on June 16, 2017
1. Abbreviations used in this paper: E-C coupling, excitation-contraction
coupling: RyR, ryanodine receptor: SR, sarcoplasmic reticulum.
chanically coupled. A conformational change in the plasma
membrane channels associated with gating-charge movement directly evokes Ca 2+ release from RyRs by proteinprotein interaction (Rios et al., 1991; Yano et al., 1995). In
some vertebrate skeletal muscle more than one RyR isoform (and both E-C coupling mechanisms) are present in
the same cells (Airey et al., 1990; Giannini et al., 1995; Jacquemond et al., 1991; O'Brien et al., 1995; Oyamada et al.,
1994).
RyRs belong to a family of intracellular Ca 2+ channels
that includes RyRs and inositol trisphosphate receptors
(Berridge, 1993). Functional RyR channels are homotetramers composed of subunits of ~565 kD. The channel
pore resides in the carboxyl 10-20% of the molecule (Callaway et al., 1994; Meissner et al., 1989), while the bulk of
the protein extends into the cytoplasm from the SR membrane (Marty et al., 1994; Radermacher et al., 1994). RyRs
have been cloned from a variety of vertebrate and invertebrate species, and are found in a number of tissues in addition to muscle (Giannini et al., 1995; Hakamata et al., 1992;
Otsu et al., 1990; Oyamada et al., 1994; Takeshima et al.,
1994b, 1989). Sequence analysis of the cloned genes (each
about 5,000 amino acids) shows extensive sequence and
structural homology, particularly in predicted transmembrane-spanning domains in the carboxyl ends of the proteins (Coronado et al., 1994; McPherson and Campbell,
1993a; Meissner, 1994).
We are investigating the role of RyRs in E-C coupling in
C. elegans striated muscle. Nematodes have relatively simple body-wall musculature used for locomotion, consisting
of rows of longitudinal striated muscle cells on dorsal and
ventral sides of the animal (Wood, 1988). Opposing dorsal
and ventral muscle cells are coordinately contracted and
relaxed, producing sinusoidal waves that propel the animal
NTRACELLULAR calcium (Ca 2+) channels play a key
role in regulating muscle contraction. During excitation-contraction coupling (E-C coupling) in striated
muscle, two classes of Ca 2+ channels act sequentially to
cause the release of Ca 2+ ions stored within the sarcoplasmic reticulum (SR) 1 (Caterall, 1991). Cholinergic excitation opens voltage-gated C a 2+ channels in the plasma
membrane, which in turn cause voltage-insensitive channels in the SR to flood the cell with C a 2+ ions, thus eliciting sarcomere contraction (Coronado et al., 1994; McPherson and Campbell, 1993a; Meissner, 1994). These high
conductance calcium release channels in the SR are
termed ryanodine receptors (RyRs) because of their specific, high affinity interaction with the plant alkaloid ryanodine (Fill and Coronado, 1988). Ryanodine locks the
channels open in a submaximal conductance state, causing
a steady leak of Ca 2+ into the myoplasm and inducing hypercontractive paralysis in skeletal muscle fibers (Fill and
Coronado, 1988).
Different striated muscle types express different RyR
isoforms, and the mechanism of E-C coupling varies in
these tissues. In cardiac muscle, external Ca > admitted
through voltage-gated channels in the plasma membrane
t r i g g e r s C a 2+ release from the SR by binding to regulatory
sites on RyRs. This mechanism is called "calcium-induced
calcium release" (CICR) (Fabiato, 1983). In skeletal muscle, the voltage-gated channels and RyRs appear to be me-
Published August 15, 1996
elegans.
Materials and Methods
Strains and Handling of C. elegans
Handling of C. elegans was done using established methods (Wood, 1988).
The strains and relevant mutations used were N2 (wild type), MT3126
(mut-2(r459)), BC1283 (sDf30), MT2583 (nD)(t2), BC1999 (nDfI8), unc68 (e540), unc-70 (e524), and dpy-ll (e224). uric-68 alleles isolated in this
study were rl l51::Tcl, rl152::Tcl , rl158, rl160, rl161, r1162, r1167, r1207,
r1208, rl209, r1210, r1211, r1212, rDfl, and rDf2. The strain TR2267 is homozygous for rl161, and contains a stable extrachromosomal array
(rEx95) containing wild-type ryr-I sequences (see below).
Molecular Methods
Preparation of nematode genomic DNA was as previously described
(Rushforth et al., 1993). Standard methods were used for preparation of
probes, for DNA sequencing, and for hybridization analysis. Screening for
Tcl insertions and deletions was done as described (Rushforth et aL, 1993;
Zwaal et al., 1993) with some modifications. Populations founded with
about 30 MT3126 animals were grown until F2 generation eggs were being
laid, and then screened in two-dimensional matrixes. Each population was
divided into four parts: two parts were used for the matrix, one part was
saved for subdivision, and one part was saved to verify a suspected insertion or deletion. For example, each of 49 populations were assigned a row
and a column in a 7 × 7 matrix. Worms from seven populations were
pooled for each row or column. DNA prepared from pooled animals in
rows and columns was screened with PCR in duplicate. A population having the same polymorphism in both its row and column was further tested
by PCR with a third portion of the animals. If positive, the remaining ani-
The Journal of Cell Biology, Volume 134, 1996
reals were subdivided into smaller populations and rescreened to ultimately isolate individual animals with insertion or deletion. Single animal
PCR was done as described (Williams et al., t 992).
Wild-type ryr-1 sequences used for transformation rescue were obtained by amplifying three 10.3-10.7-kb fragments from N2 genomic DNA
with the Expand r~ long template PCR system (Boehringer Mannheim
Corp., Indianapolis, IN). The three ryr-I primer pair sequence coordinates (5'base and length of primer) are 1,520 (32-mer, sense direction)
and 11,870 (32-mer, antisense); ll,440 (39-mer, sense), and 21,770 (35mer, antisense); 21,361 (25-mer, sense) and 32,145 (35-mer, antisense).
The PCR fragments were purified on spin columns, mixed 1:25 with linear
DNA fragments (B RL-l-kb ladder), and injected into unc-68(r1161) adult
hermaphrodites as described (Mello and Fire, 1995) at a total DNA concentration of 100 ~g/ml. Transformants were identified by screening the
F2 progeny of injected (Unc) animals for animals with wild-type motility.
Isolation of unc-68(rl l58)
uric-68 (r1158) was obtained by screening populations of rll51::Tcl with
flanking primers as described (Zwaal et al., 1993). A 5.4-kb deletion of
ryr-l, contained in strain TR2125, was identified in this screen. This deletion, r1158, removes bps 24,530 to 29,902 of ryr-l. TR2125 is phenotypically wild type, but proved to contain a duplication of ryr-l. In one copy of
ryr-l, TR2125 is homozygous for r1158. In a second copy, TR2125 is homozygous for rll51::Tcl. Both the deletion-containing and insertion-containing alleles of ryr-I are tightly linked to unc-70 (data not shown, see
Fig. 1). During these tests of linkage, LGV from TR2125 was placed in
trans to unc-70(e524). From one such heterozygote, animals having an
Unc phenotype distinct from that of unc-70 segregated unexpectedly.
Southern analysis and PCR tests demonstrated that these novel Unc animals (strain TR2128) were homozygous for r1158 but no longer contained
rllSl::Tcl. We have not characterized further the duplication/deletion rearrangement of TR2125, but we presume that the simple deletion r1158 in
TR2128 represents resolution of the duplication via homologous recombination.
Noncompleraentation Screens
In the screen for new spontaneous alleles, wild-type males were crossed to
r1151:.'Tcl or rl152::Tc1 hermaphrodites. The resulting ryr-I::Tcl/+ males
were then crossed to dpy-ll(e22) unc-68(rl158) homozygotes. The F1
progeny were then screened for Unc non-Dpy hermaphrodites. Putative
new unc-68 alleles were outcrossed, tested for eomplementation with unc68(e540), and checked for the absence of the r1158 deletion with PCR.
New alleles were characterized with PCR and Southern analysis. In the
screen for EMS-induced alleles, wild-type males were mutagenized with
EMS (Anderson, 1995) and crossed with dpy-ll(e224) unc-68(rl158) hermaphrodites as above. Putative new EMS-induced alleles were outcrossed, tested for complementation with unc-68(e540), and checked for
the absence of the r1158 deletion with PCR. Six new alleles were designated r1207-r1212.
Binding Assays and Membrane Purification
Whole C. elegans were collected, floated on 35% sucrose (Wood, 1988)
and washed in H20. Washed worms were resuspended in 20 mM MOPS,
pH 7.2, in a protease inhibitor cocktail previously described (Kim et al.,
1992). Worms were homogenized by passing twice through a french pressure cell at 10,000 psi. To purify the microsomal fraction, the crude homogenates were spun at 2,400 g for 30 rain in corex tubes. The supernatant
was layered on an equal volume of 30% sucrose (in 0.4 m KC1, 20 mM
MOPS, pH 7.2, pCa 4.3) and spun at 90,000 g for 45 min. The pellets were
resuspended in 1.0 M KCI, 20 mM Tris, pH 8.0, pCa 4.3. Binding assays
were done in resuspension buffer at 37°C with indicated concentrations of
[3H]ryanodine (Dupont-New England Nuclear, Boston, MA). Assays
were in triplicate, with 70 Ixg microsomal protein per sample. Bound
[3H]ryanodine was collected on glass fiber filters and counted in scintillation cocktail as described (Kim et al., 1992). Specific counts for each ryanodine concentration were determined by subtracting counts bound in the
presence of 10l/ p,M unlabeled ryanodine (Agrisystems International,
Wind Gap PA) from counts in the presence of labeled ryanodine only.
Electron Microscopy
Animals were fixed for 1 h on ice in a solution composed of 0.67% fresh
glutaraldehyde and 0.67% osmium tetroxide in 0.1 M cacodylate buffer,
886
Downloaded from on June 16, 2017
forward or backward (Nicholas, 1984). Electrophysiological experiments in Ascaris lumbricodes have established
that the contractile state of body-wall muscle is controlled
by membrane potential (Weisblat et al., 1976; Johnson and
Stretton, 1980), which itself is controlled by excitatory and
inhibitory inputs from motor neurons (Walrond et al.,
1985). Excitatory cholinergic and inhibitory GABAergic
neuromuscular synapses have been characterized in Ascaris and C. elegans (Johnson and Stretton, 1980; Lewis
et al., 1987; Mclntire et al., 1993), but nothing is known
about E-C coupling mechanisms that directly control cytoplasmic Ca ~+ levels (and therefore contraction) in nematodes. The location and nature of the nematode SR is not
certain; it is assumed to consist of vesicular structures that
surround dense bodies, the attachment sites for thin filaments in nematode sarcomeres (Nicholas, 1984; Wood, 1988).
C. elegans is known to have RyRs. We characterized a
[3H]ryanodine-binding activity in C. elegans membranes
that copurifies with high conductance ion channels. The
channels are cation and Ca 2+ selective, and are gated by
ryanodine (Kim et al., 1992). Addition of ryanodine to the
growth medium induces an incomplete hypercontractive
paralysis of C. elegans, suggesting an effect on body-wall
muscle (Kim et al., 1992). Sakube and Kagawal have
cloned, sequenced, and characterized a RyR gene in C. elegans. This gene, called ryr-I, is predicted to encode a protein of 5,017 amino acids that is ~40% identical to cloned
RyRs from vertebrates. Ten membrane-spanning domains
are predicted in the carboxyl end of ryr-1, the region of
greatest homology to mammalian RyRs (Sakube et al.,
1993). ryr-I promoter-[3-galactosidase fusion constructs
are expressed in adult body-wall muscle and pharyngeal
muscle ceils (Kagawa, H., personal communication). To
better understand the role of RyRs in E-C coupling in
nematodes, we undertook a genetic analysis of ryr-1 in C.
Published August 15, 1996
pH 7.4 (C buffer). The heads were cut off, the animals washed three times
in C buffer, and then incubated at 4°C overnight in 2% osmium tetroxide
in C buffer. Animals were washed again, dehydrated, and embedded in
Spurr's resin. Adult hermaphrodites were sectioned perpendicular to the
long axis of the body posterior to the terminal bulb of the pharynx. The
sections were viewed in a JEOL (Peabody, MA) 100 electron microscope
at a magnification 19,000.
Results
Genetic and Physical Location of ryr-1
A C. elegans ryanodine receptor homologue c D N A (cm16C2)
was identified in a screen of expressed genes (Waterston
et al., 1992). The sequence of cm16C2, a 1,232-bp partial
c D N A of ryr-1, is homologous over its entire length to several mammalian ryanodine receptors in the databases
( ~ 3 0 % identical), cm16C2 hybridized to three overlapping Y A C s on linkage group V (see Fig. 1). These Y A C s
span the interval between snb-1 and odr-2, two genes
whose positions are known on both the physical and genetic maps (see Fig. 1). We located cm16C2 more precisely
on the genetic map using several overlapping deficiencies
affecting this region. Deficiency homozygote embryos
were tested with P C R for cm16C2 sequences, cml6C2
(ryr-1) mapped to an interval of ~0.1 map units bounded
by the right breakpoints of deficiencies nDf32 and sDf20
(see Fig.l, shaded area). This interval contains unc-68 and
odr-2, and eight complementation groups defined by recessive lethal mutations (Johnson and Baillie, 1991).
Site-selected Tc l Insertions within ryr-1
Isolation o f ryr-1 Deletions
Figure 1. Physical and genetic maps of LGV surrounding ryr-1.
The partial ryr-1 cDNA cm16c2 hybridized to three (bold type) of
four overlapping YACs shown in the expanded area. ryr-1 is ordered on the YAC contig with respect to snb-1 and odr-2, two
genes whose location is known on both genetic and physical
maps. Cosmids having all or a portion of snb-1, odr-2, or ryr-I are
shown with gene names below the contig, snb-1 and odr-2 are
also ordered on the genetic map above with respect to dpy-11,
unc-70, unc-68, and three deficiencies: nDfl8, nDf32, and sDf30.
The shaded area shows the interval to which the cm16C2 clone
mapped. The relative order of snb-1 and unc-70 has recently been
established (Nonet, M., and E. Jorgensen, personal communication).
Because Tcl sequences are spliced from most, if not all
Tcl-containing p r e - m R N A s (Rushforth and Anderson,
1996), we suspected that the wild-type phenotypes of
r1151 and r1152 did not represent the ryr-1 null phenotype. Using previously described methods (Zwaal et al.,
1993), we isolated r1158, an excision derivative of r1151::
Tcl, in which 5.4 kb of ryr-1 are deleted (see Fig. 2 and 3,
lane 3). Sequencing of the r1158 deletion junction (generated by PCR) demonstrated that ryr-1 coordinates 24,531
to 29,901 (inclusive) are deleted in r1158, r1158 homozygotes are uncoordinated and exhibit a phenotype that is
identical to that of unc-68(e540), the unc-68 reference allele. Furthermore, r1158 fails to complement unc-68(e540).
To confirm that the Unc-68 phenotype of r1158 was
caused by the ryr-1 deletion, a noncomplementation screen
for new alleles of unc-68 was performed, using rll51::Tcl
and rl152::Tcl as a source of new mutations. O u r expectation was that imprecise excision of Tcl in a mutator background would yield additional deletion alleles of ryr-1 at
relatively high frequency (Zwaal et al., 1993). If the Unc68 phenotype of r1158 were due to a mutation in a gene
other than ryr-1, then the frequency of new unc-68 alleles
would be very low, or the deletions obtained from Tcl excisions would always extend beyond ryr-1 and include
nearby genes. Six new spontaneous alleles of unc-68 were
isolated from 2,330 rll51::Tcl and r1152::Tc1 chromosomes screened (see Materials and Methods). Three of
these alleles (r1160, r1161, and r1162) delete the T c l element used to generate them as well as 2.4-7.2 kb of adjacent ryr-1 sequences (see Fig. 2, also see Fig. 3, lanes 4, 6,
and 9). Two alleles (rDfl and rDf2) contain deletions that,
based on P C R tests, extend beyond ryr-1 coding sequences. Both rDfl and rDf2 are lethal as homozygotes,
Maryonet at. unc-68Encodesa C. elegansRyanodineReceptor
887
Downloaded from on June 16, 2017
Y. Sakube and H. Kagawa independently identified ryr-1,
and determined the complete sequence of both genomic
and c D N A clones, ryr-1 is composed of 46 exons spanning
over 30-kb of genomic D N A (see Fig. 2). With genomic sequence information generously provided by Sakube and
Kagawa, we screened for animals having Tcl transposable
element insertions within the 3' third of ryr-1. Using a
PCR/sib-setection method previously described (Rushforth et al., 1993), two site-selected Tcl insertions were obtained (see Fig. 2). Sequencing of P C R fragments spanning the insertion junctions revealed that r1151 contained
an insertion within intron 37 at T A dinucleotide 27,14727,148 (coordinates of Sakube et al., Genbank accession
number D54899) and r1152 contained an insertion within
the 5' splice donor of intron 43 ( T A dinucleotide 29,618-29,619). Southern hybridization of D N A from insertion
homozygotes confirmed the size and location of the Tcl
insertions in r1151 and r1152 (Fig. 3, lanes 2 and 9, respectively). Both r1151 and r1152 homozygotes are wild-type
in phenotype.
Published August 15, 1996
ryr-1
Genomic
Region:
Exons
20-46
Figure 2. Tcl site-selected
insertions and Tcl-induced
deletions. Exons 20-46 of ryr-1
showing the location of two
Tcl insertions and four deler1151::Tcl
r1152::Tcl
tions, r1158 was isolated
from rll51::Tcl using the
5
'
~
"
3'
primer pair shown (see MaTAA
terials and Methods). Three
alleles isolated in a noncom--i~[
r1158 5.4 kb
I
plementation screen (rl160,
r1161, r1162, see Materials
and Methods) are deletions
[
r1162 3.7 kb
J
within ryr-1 that extended
unidirectionally from the Tcl
I
r1161 7.2 kb
I
insertion from which they
were derived, r1161 and
r1162 were derived from
[ r1160 2.4 kb 1
r1151, while r1160 was derived from r1152. The sizes of
I = predicted transmembrane domain
r1158, rl160, and r1162 were
1 kb
determined
by sequencing
-1~ = PCR primers
deletion junction fragments.
The size of rl161 was estimated from Southern analysis and from the size of PCR fragments amplified from primers flanking the deletion. Ten predicted transmembrane-spanning domains are shown as vertical black bars (Sakube et al., 1993).
that unc-68 and ryr-1 are the same gene. Hereafter, we refer to this gene simply as unc-68.
Rescue of unc-68 with ryr-1 Genomic D N A
Three lines of evidence indicate that the viable, uncoordinated phenotype of the unc-68 deletion homozygotes represents the unc-68 null phenotype. First, the four deletions
remove substantial (and differing) portions of the uric-68
coding sequence, yet exhibit the same Unc-68 phenotypes.
In particular, r1162 deletes all of the predicted membranespanning domains (Sakube et al., 1993) (see Fig. 2). Since
homologous domains in mammalian RyRs appear to contain the channel pore (Callaway et al., 1994, Meissner et al.,
1989) r1162 is certainly a "channel null" allele, regardless
of whether it is a protein null allele. Second, the deletion
mutations are fully recessive, and their Unc-68 phenotype
is not more severe when any of them is placed in trans to a
large deficiency that removes all of uric-68 (sDf20, see Fig.
1). Third, new alleles of unc-68 that are indistinguishable
from those described above arise at high frequency following EMS mutagenesis. From 2871 EMS-mutagenized chromosomes, we isolated six new uric-68 alleles (see Materials
and Methods). Such frequencies of isolation suggest that
the Unc-68 phenotype is due to loss of function alleles
(Anderson, 1995).
As further proof that unc-68 and ryr-I are synonymous,
ryr-1 genomic D N A was used for transformation rescue of
unc-68 mutants (Mello and Fire, 1995). The cosmid
M04Cll (see Fig. 1) contains the entire ryr-1 coding sequence, but M04Cll D N A failed to rescue the Unc-68
phenotype. Y. Sakube and H. Kagawa have shown that
body-wall muscle expression of ryr-1 requires noncoding
upstream regulatory sequences not present in M04Cll
(Kagawa, H., personal communication). We therefore
used PCR to amplify ryr-1 and 3 kb of upstream sequence
from N2 genomic D N A in three 10.3-10.7-kb fragments,
which together span ryr-1 sequence coordinates 1,52032,145 (see Materials and Methods). The central fragment
overlapped the 5' and 3' fragments by 300-350 bp to allow
homologous recombination in vivo (see Mello and Fire,
1995).
The three PCR fragments were injected into unc68(r1161) hermaphrodites. F2 progeny of the injected Unc
animals included wild-type hermaphrodites. One such animal established the strain TR2267, which carries an extrachromosomal array (rEx95) that includes unc-68(+). TR2267
wild-type hermaphrodites segregate wild-type and Unc
progeny. PCR of TR2267 wild-type animals showed that
both the r1161 deletion and wild-type alleles of ryr-1 were
present, while TR2267 Unc animals contained only the
r1161 deletion allele, rex95 was also able to rescue other
unc-68 alleles, including e540, as shown by crossing wildtype TR2267 males to unc-68 homozygotes and recovering
wild-type F1 progeny. These experiments demonstrate
The Journalof Cell Biology,Volume134, 1996
uric-68 Null Phenotype
Effects of unc-68 Mutations on Muscle Function
Wild-type C. elegans moves by propagating sinusoidal contraction waves in strips of striated muscle cells attached to
the cuticle along the body length (Croll, 1975; Nicholas,
1984). Contraction waves of unc-68 homozygotes propagate more slowly than those of wild type. unc-68 mutants
move slowly, and exhibit a languid, incomplete flaccid paralysis. The animals travel for shorter average distances
888
Downloaded from on June 16, 2017
presumably because they delete other genes in addition to
ryr-1. A sixth allele (r1167, see Fig. 3, lane 10) retains the
rl152::TcI insertion. It is being further characterized to
determine the nature of its new mutation.
Published August 15, 1996
than wild type, are immobile for more extended periods of
time, and assume unusual extended or curled body postures. In a motility assay performed in liquid media, unc68 mutants propagated 25-30% the number of contraction
waves observed in wild-type animals (see Table I). As
shown in Table I, the motility of unc-68(rl161) was restored to wild-type level by rescue with the unc-68(+) array (see rl161;rEx95).
unc-68 mutants do not exhibit defects of anal depressor
or sex muscles, although pharyngeal muscle function may
be affected. Defecation is grossly normal (Reiner, D., and
J. Thomas, personal communication), and unc-68 mutants
are proficient in egg laying. While unc-68 males mate
poorly, this is most likely a consequence of their impaired
motility. Pharyngeal pumping is somewhat weaker in unc68 mutants than in wild-type, although electrical activity
during pharyngeal muscle contraction is normal in unc-68
mutants (Lee, R., and L. Avery, personal communication).
The mutants grow more slowly than their wild-type counterparts, perhaps due to weaker pumping (feeding) during
larval development.
Body-Wall Muscle Ultrastructure Is Normal in
unc-68 Mutants
Figure 3. Southern hybridization analysis of Tcl insertion and
deletion mutants. Genomic DNA from wild-type or unc-68 mutants was digested with BgllI (G) + SpeI (S), or HindlII (H). Filter strips from the same gel were hybridized to probe A or B, as
indicated above the lanes. The sizes of the hybridizing bands are
shown in kilobases at left for BgllI + SpeI digests, or at right for
HindlII digests. Restriction sites and fragment sizes of N2 DNA
(wild type), and the location of probes A and B are shown on the
map above. Also indicated are the locations of the Tel insertions.
The map shows ryr-1 (unc-68) sequences from the BgllI site at bp
19,263, (exon 19) to the HindlII site at bp 34,321, 2.2 kb 3' of the
poly-A addition site (Sakube et al., 1993).
unc-68 Mutants Are Insensitive to Ryanodine
Contraction of body-wall muscle is impaired, but not eliminated in unc-68 mutants. Because many species express
Table I. Motility Assay for Wild-type and unc-68 Mutant Animals
unc-68 allele
W.T.
e540
r1158
rl160
r1162
r1161
r1161; rex95
-- R y a n o d i n e
exp. 1
exp. 2
107 ± 8
114 ± 18
29 ± 4.6
28 +-- 3.8
29 ± 5.5
28 - 5.0
27 + 4 . 4
27 ± 4.1
22 ± 4.0
22 ± 5.0
25 ± 4.5
24 --- 4.4
103 ± 8.1
104 _ 8.6
+ Ryanodine
exp. l
exp. 2
12 ± 1.9
12 -+ 2.6
30 ± 4.6
28 ± 6.3
29 -+ 5.5
28 - 3.5
28 ± 7.5
27 ± 5,3
23 +_ 4.9
23 ± 5.3
25 ± 4.7
23 ± 3.7
8 +_ 1.6
9 ± 1.6
The relative motility of wild-type and unc-68 mutants was measured by counting the waveforms (Croll, 1975) propagated by individual animals in one minute. Each value is the
average number of waveforms counted (-+ SD) in ten animals. Before counting, adult hermaphrodites were placed in 24-well plates in 0.5 ml S-media (Wood, 1988) with or without 2 ~M ryanodine. The plates were incubated at room temperature, while shaking, until wild-type animals were fully contracted by ryanodine (at least 30 rain). A waveform was
defined as the propagation of a contraction wave from head-to-tail, from tail-to-head, or the curling and uncurling of the body. Results from two independent experiments are
shown.
Maryon et al. unc-68 Encodes a C. elegans Ryanodine Receptor
889
Downloaded from on June 16, 2017
The motility defects of unc-68 mutants could arise from
structural abnormalities of the muscle contractile apparatus, or from defective regulation of contraction. We compared wild-type and unc-68 body-wall muscle using electron
microscopy, unc-68 sarcomeres exhibit nearly wild-type ultrastructure, as seen in transverse sections (see Fig. 4). The
distribution of thick and thin filaments, and the number
and location of dense bodies and M-lines (Wood, 1988)
were essentially normal in unc-68 animals. We observe
two subtle irregularities in unc-68 body-wall muscles. First,
the depth of the spindle region (but not its myofilament
organization) is more heterogeneous. Thus, many unc-68
sarcomeres are deeper than their wild-type counterparts.
Second, dense bodies are more often irregular in shape
than wild type (e.g., lower left image in Fig. 4). We conclude from this analysis that unc-68 body-wall muscle ultrastructure is nearly normal, and that motility defects are
due to defective regulation of contraction.
Published August 15, 1996
Downloaded from on June 16, 2017
Figure 4. Transverse sections of wild-type (N2) and unc-68 body-wall muscle. Wild type or unc-68 (r1162) animals were prepared for
EM as described in Materials and Methods. Transverse sections from adult hermaphrodites taken posterior to the terminal bulb of the
pharynx are shown. Magnification, 19,000. db, dense bodies; m, M line (Nicholas, 1984; Wood, 1988) unc-68 (r1162) exhibits nearly normal muscle ultrastructure. Bar, 1.0 txM.
two R y R isoforms in the same muscle cell type (Airey et al.,
1990; Giannini et al., 1995; O'Brien et al., 1993; O y a m a d a
et al., 1994), we tested unc-68 mutants for sensitivity to
contraction by ryanodine, which induces incomplete hypercontractive paralysis of C. elegans (Kim et al., 1992).
After exposure to ryanodine, wild-type animals become
shorter and thicker, indicating that the body-wall muscles
are hypercontracted. Wild-type animals in 2 mM ryanodine move very slowly compared to animals without drug,
as shown in a motility assay (see Table I), and by the
length of tracks left in bacterial lawns (Fig. 5). In contrast,
unc-68 mutants are completely resistant to ryanodine.
The Journal of Cell Biology, Volume 134, 1996
890
Published August 15, 1996
Both the unc-68 deletion mutants described above and the
unc-68 reference allele e540 are unaffected by 2 mM ryanodine, a concentration 20 times higher than the minimum
required to paralyze wild type (see Table I and Fig. 5).
Rescue of unc-68(rl161) with the wild-type unc-68 gene
restored sensitivity to ryanodine (see rl161; rEx95, Table I).
These experiments show that unc-68-encoded RyRs alone
are responsible for the effects of ryanodine on body-wall
muscle function.
Discussion
Microsomes prepared from C. elegans homogenates contain a saturable, high affinity [3H]ryanodine-binding activity. Although the amount of binding activity is low in comparison to rabbit skeletal muscle, the binding affinity (Kd
26 nM) is similar to that measured for vertebrate RyRs
(Kim et al., 1992). If unc-68 encodes the only RyR in C. elegans (as defined by ryanodine binding), then unc-68 mutants should lack high affinity ryanodine-binding activity.
We measured [3H]ryanodine binding in microsomes prepared in parallel from wild-type or unc-68 mutants. One
such experiment is shown in Fig. 6. We were unable to detect specific [3H]ryanodine binding in unc-68 microsomes,
while those from wild-type animals contained saturable,
high affinity [3H]ryanodine-binding sites. These data, together with the ryanodine resistance of unc-68 mutants,
suggest that uric-68 encodes the only RyR in C. elegans.
We provide five lines of evidence demonstrating that unc68 encodes a ryanodine receptor (RyR) involved in regulating C. elegans body-wall muscle contraction. First, the
cloned RyR gene ryr-1 (Sakube et al., 1993) maps to a genetic interval that includes uric-68, and intragenic deletions of ryr-1 isolated with reverse genetic methods and
noncomplementation screens are alleles of unc-68. Second, transformation of unc-68 mutants with ryr-1 sequences fully rescues the Unc-68 phenotype. Third, unc-68
mutants exhibit motility defects (incomplete paralysis),
but their body-waU muscle ultrastructure is essentially normal. Fourth, unc-68 mutants are insensitive to the paralyzing effects of ryanodine. Fifth, microsomal fractions of
unc-68 mutants are devoid of the high affinity binding activity that characterizes wild-type microsomes.
Sakube and Kagawa have shown that unc-68 is expressed in body-wall muscle cells (Kagawa, H., personal
communication), ryr-1 promoter fusions to lacZ are expressed in body-wall muscle cells, and are variably expressed
in pharyngeal muscle and in egg-laying muscles. We and
others have not observed severe behavioral defects of uric68 mutants in pharyngeal pumping or egg laying, although
the weaker pumping (Lee, R., and L. Avery, personal
communication) and slow larval growth (Maryon, E., unpublished results) of unc-68 mutants suggest a role of unc68 in the regulation of pharyngeal muscle contraction.
Maryon et al. unc-68 Encodes a C. elegans Ryanodine Receptor
891
unc-68 Mutants Lack Ryanodine-binding Activity
Downloaded from on June 16, 2017
Figure 5. Effect of ryanodine on the motility of wild-type or unc-68 animals. Five adult hermaphrodites were placed on 35-mm agar
plates having bacterial lawns. Ryanodine-treated animals were first incubated in M9 media (Wood, 1988) containing 2 I~M ryanodine for
30 min (a time sufficient to maximally paralyze wild type), and then placed on agar plates containing 2 p,M ryanodine. After 10 min, the
plates were photographed through a Wild M3Z dissecting microscope equipped with a phototube. In the picture showing wild-type
without ryanodine (top left), the animals made extensive tracks covering the plate, only some of which are visible in the field.
Published August 15, 1996
70-
The Journal of Cell Biology, Volume 134, 1996
892
,~
50-
20-
WildType
S
unc-ea(r116e)
I
0
10
20
nM
30
40
50
60
3H-ryanodine
Figure 6. [3H]Ryanodine binding to C. elegans microsomes. Microsomes were prepared from either wild-type or unc-68 (r1162)
mutants as described in Materials and Methods. After incubation
for 2 h at 37°C in [3H]ryanodine at indicated concentrations, the
bound counts were collected on glass fiber filters. Samples were
assayed in triplicate for specific and nonspecific binding. Specific
binding was calculated as average [3H]ryanodine counts bound at
a given drug concentration in excess of the counts bound in the
presence of 100 ixM unlabeled ryanodine. The calculated Kd of
[3H]ryanodine binding to wild-type microsomes was 24 nM, close
to the value of 26 nM obtained in earlier studies (Kim et al.,
1992).
Downloaded from on June 16, 2017
RyRs play an important, but nonessential role in E-C
coupling in C. elegans body-wall muscle, unc-68 appears to
encode the only RyR expressed in C. elegans, since unc-68
mutants are resistant to the paralytic effects of ryanodine
and lack detectable [3H]ryanodine-binding activity. Based
on these observations, RyRs are not necessary for contraction of body-wall muscle because uric-68 mutants propagate attenuated sinusoidal contraction waves up or down
the body length (see Table I and Fig. 5). Furthermore, in
wild-type animals, ryanodine induces an incomplete paralysis. At high drug concentration, the hypercontracted animals still move with slow, sinusoidal motion, as shown by
the tracks left in bacterial lawns (see Fig. 5). Thus, coordinated contraction and relaxation of body-wall muscle cells
still occurs with either a (presumed) deficit of RyR-supplied Ca 2÷ in unc-68 mutants or an excess of Ca 2+ in the
presence of ryanodine. Wild-type RyR function greatly
enhances motility (see Table I and Fig. 5), but is not necessary for E-C coupling per se.
The RyR-independent contraction observed in unc-68
body-wall muscle cells could be initiated by Ca 2÷ from an
external or internal source. In many invertebrates, including nematodes, muscle plasma membrane depolarization
emanates from neuromuscular junctions as a graded Ca 2+
spike (Hagiwara and Naka, 1964; Weisblat et al., 1976). In
C. elegans, Ca 2+ entering through plasma membrane voltage-gated channels might suffice to initiate contraction,
unlike the modest Ca 2+ flux through analogous channels
in cardiac myocytes (Varro et al., 1993). The small size of
body-wall muscle cells and close proximity of sarcomeres
to the plasma membrane could reduce the need for internal Ca 2+ release. On the other hand, an internal Ca 2+
source cannot be ruled out. For example, cardiac myocytes
and smooth muscle cells contain inositol trisphosphategated Ca 2+ stores (Otani et al., 1988; Somlyo and Somlyo,
1994). Regardless of the source of Ca 2+ that evokes RyRindependent contraction in unc-68 mutants, RyRs appear
to amplify the signal in body-wall muscle cells.
Certain R y R isoforms in vertebrate skeletal muscle also
appear to amplify calcium transients. The major skeletaltype RyRs, termed RYR-1 in mammals or a - R y R in other
vertebrates (Airey et al., 1990; McPherson and Campbell,
1993a), are directly gated by membrane potential (Rios
et al., 1991). A second skeletal isoform, termed RYR-3 in
mammals or 13-RyR in other vertebrates, does not respond
directly to depolarization (Airey et al., 1990, McPherson
and Campbell, 1993a; Oyamada et al., 1994). Mutations
that specifically eliminate RYR-1 channels in mice
(skrr ml) or a-RyRs in chickens (cn) have recessive lethal
phenotypes, resulting in perinatal death (Ivanenko et al.,
1995; Takeshima et al., 1994a). Muscle fibers from mutant
embryos tack normal Ca 2÷ transients or contractile responses upon electrical stimulation, but exhibit these responses when exposed to caffeine or treatments that increase Ca 2÷ entry through plasma membrane channels
(Ivanenko et al., 1995; Takeshima et al., 1994a). Further
experiments suggest that the responses of mutant fibers
are due to CICR through RyR-3 or 13-RyR-type channels
(Percival et al., 1994; Takeshima et al., 1995). These and
other experiments have provoked a two receptor model
for Ca 2+ release in skeletal muscle, in which [3-type RyRs
amplify the Ca 2+ released by a-type RyRs by CICR (Ivanenko et al., 1995; Jacquemond et al., 1991; O'Brien et al.,
1995).
The amplification of Ca 2+ signals by CICR is probably
the more ancient and most common role of RyRs in E-C
coupling. The major skeletal isoforms (and "mechanical"
E-C coupling) appear to have evolved from [3-RyR-type
channels to serve the needs of fast twitch muscle types
(O'Brien et al., 1993). Other RyR isoforms in cardiac muscle, in smooth muscle, in crustaceans, and in the CNS are
known (or believed) to use a CICR mechanism (Fabiato,
1983; Gy6rke and Palade, 1992; Lynn and Gillespie, 1995;
McPherson and Campbell, 1993b). We infer from the Unc68 phenotype that RyRs in nematodes, one of the more
primitive metazoan phyla, also amplify Ca 2+ signals. While
there is no direct evidence of a CICR mechanism for nematode RyRs, it is notable that UNC-68 is more similar to
cardiac RyRs than to the major skeletal isoforms (Sakube
et al., 1993).
Our experiments lead us to conclude that the role of
RyRs in C. elegans body-wall muscle is to enhance contraction by amplifying a depolarization-coupled Ca 2+ transient. We consider it less likely that RyRs are involved in
relaxation (e.g., repolarization of the muscle membrane),
as proposed for RyRs in arterial smooth muscle ceils (Nelson et al., 1995), because the absence of RyR function in
unc-68 mutants results in languid, flaccid paralysis, while
ryanodine, known to lock the channels open, causes hypercontraction of wild-type animals. Further studies of unc68-encoded RyR channels will help to uncover the molecular mechanisms regulating nematode body-wall muscle
contraction, and should also be useful in understanding
the evolution of RyRs and E-C coupling in more complex
organisms.
Published August 15, 1996
Maryon et al. unc-68 Encodes a C. elegans Ryanodine Receptor
893
Received for publication 28 March 1996 and in revised form 18 June 1996.
References
Downloaded from on June 16, 2017
Airey, J.A., C.F. Beck, K Murakami, S,J. Tanksley, T.J. Deerinck, H. Ellisman,
and J.L. Sutko. 1990. Identification and localization of two triad junctional
foot protein isoforms in mature avian fast twitch skeletal muscle. J. Biol.
Chem. 265:14187-14194.
Anderson, P. 1995. Mutagenesis. In Methods in Cell Biology. Vol. 48. H.E. Epstein and D.C. Shakes, editors. Academic Press, San Diego. 31-54.
Berridge, M.J. 1993. lnositol triphosphate and calcium signalling. Nature
(Lond.). 361:315-325.
Callaway, C., A. Seryshev, J.P, Wang, K,J. Slavik, D.H. Needleman, C. Cantu,
Y. Wu, T. Jayaraman, A.R. Marks, and S.L. Hamilton. 1994. Localization of
the high and low affinity [3H]ryanodine binding sites on the skeletal muscle
Ca + release channel. Z BioL Chem. 280:15876-15884.
Caterall, W.A. 1991. Excitation-contraction coupling in vertebrate skeletal
muscle: a tale of two calcium channels. Cell, 64:871-874.
Coronado, R., J. Morrissette, M. Sukhareva, and D.M. Vaughan. 1994. Structure and function of ryanodine receptors. Am. J, PhysioL 94:1485-1504.
Croll, N.A. 1975. Components and patterns in the behavior of the nematode
Caenorhabditis elegans. J. Zool, Lond. 176:t59-176.
Fabiato, A. 1983. Calcium-induced release of calcium from the cardiac sarcoplasmic reticulum. Am. J. Physiol. 245:C1-C14.
Fill, M., and R. Coronado. 1988. Ryanodine receptor channel of sarcoplasmic
reticulum. Trends Neurosci. 11:453-457.
Giannini, G., A. Conti, S. Mammarella, M. Scrobogna, and V. Sorrentino. 1995.
The ryanodine receptor/calcium channel genes are widely and differentially
expressed in murine brain and peripheral tissues. J. Cell Biol, 128:893-904.
Gy6rke, S., and P. Palade. 1992. Calcium-induced calcium release in crayfish
skeletal muscle. Z Physiol. 457:195-210.
Hagiwara, S., and K.-I. Naka. 1964. The initiation of spike potential in barnacle
muscle fibers under low intracellular Ca ++. Z Gen. Physiol. 48:141-162.
Hakamata, Y., J. Nakai. H. Takeshima, and K, lmoto. 1992. Primary structure
and distribution of a novel ryanodine receptor/calcium release channel from
rabbit brain, FEBS Lett. 312:229-235.
lvanenko, A., D.D. McKemy, J.L, Kenyon, J.A. Airey, and J.L. Sutko. 1995.
Embryonic chicken skeletal muscle cells fail to develop normal excitationcontraction coupling in the absence of the a-ryanodine receptor. J. BioL
Chem. 270:4220-4223.
Jacquemond, V., L. Csernoch, M.G. Klein, and M.F. Schneider. 1991. Voltagegated and calcium-gated calcium release during depolarization of skeletal
muscle fibers. Biophys. J. 60:867-873.
Johnson, C.D., and A.O.W. Stretton. 1980. Neural control of locomotion in Ascaris: anatomy, physiology, and biochemistry. In Nematodes as Biological
Models. Vol. 1. B.M. Zuckerman, editor, Academic Press, New York. 159-195.
Johnson, R.C., and D.L. Baillie. 1991. Genetic analysis of a major segment
[LGV(left)] of the genome of Caenorhabditis elegans. Genetics, 129:735-752.
Kim, Y.K,, H.H. Valdivia, E.B. Maryon, P.A. Anderson, and R. Coronado.
1992. High molecular weight proteins in the nematode C. elegans bind [3H]
ryanodine and form a large conductance channel. Biophys. J. 63:1379-1384.
Lewis, J.A., J.S. Elmer, J. Skimming, S. Mclafferty, J. Flemming, and T. McGee.
1987. Cholinergic receptor mutants of the nematode Caenorhabditis elegans.
J. Neurosci. 7:3059-3071.
Lynn, S., and J.l. Gillespie. 1995. Basic properties of a novel ryanodine-sensifive, caffeine-insensitive calcium-induced calcium release mechanism in permeabilized human vascular smooth muscle cells. FEBS Lett. 367:23-27.
Marry. I., M. Villaz, G. Arlaud, I. Bally, and M. Ronjat. 1994. Transmembrane
orientation of the N-terminal and C-terminal ends of the ryanodine receptor
in the sarcoplasmic reticulum of the rabbit skeletal muscle. Bi~phys. J. 298:
743-749.
Mclntire, S.L., E. Jorgensen, and H.R. Horvitz. 1993. Genes required for
GABA function in Caenorhabditis elegans. Nature (Loud.). 364:334-337.
McPherson, P.S., and K.P. Campbell. 1993a. The ryanodine receptor/Ca 2+ release channel, J. Biol. Chem. 268:13765-13768.
McPherson, P.S., and K.P. Campbell. 1993b. Characterization of the major
brain form of the ryanodine receptor/Ca 2+ release channel. J. Biol. Chem.
268:19785-19790.
Meissner, G. 1994. Ryanodine receptor/Ca z+ release channels and their regulation by endogenous effectors. Annu. Rev. Physiol. 56:485-508.
Meissner. G., E. Rousseau, and F.A. Lai. 1989. Structural and functional correlation of the trypsin-digested Ca 2+ release channel of skeletal muscle sarco-
plasmic reticulum. Z BioL Chem. 264:1715-1722.
Mello, C., and Fire, A. 1995. DNA transformation. In Methods in Cell Biology.
Vol. 48. H,E. Epstein and D.C. Shakes, editors. Academic Press, San Diego,
CA. pp. 452-482.
Nelson, M.T., H, Cheng, M. Rupart, L,F. Santana, A.D. Bonev~ H.J. Knot, and
W.J. Lederer. 1995. Relaxation of arterial smooth muscle by calcium sparks.
Science (Wash, DC). 270:633-636.
Nicholas, W.L. 1984. The Biology of Free Living Nematodes. Clarendon Press,
Oxford. pp. 31-53.
O'Brien, J., G. Meissner, and B. Block. 1993. The fastest contracting muscles of
nonmammalian vertebrates express only one isoform of the ryanodine receptor. Biophys. J. 65:2418-2427.
O'Brien, J., H.H. Valdiva, and B. Block. 1995. Physiological differences between the cxand 13ryanodine receptors of fish skeletal muscle. Biophys. J. 68:
417-482.
Otani, H,, H, Otani, and D. Das. t988. a-Adrenoceptor-mediated phosphoinositide breakdown and inotropic response in rat left ventricular papillary
muscles. Circ. Res. 62:8-17.
Otsu, K., H.F. Willard, V.K. Khanna, F. Zorzato, N.M. Green, and D.H. MacLennan. 1990. Molecular cloning of cDNA encoding the Ca 2+ release channel (Ryanodine Receptor) of rabbit cardiac muscle sarcoplasmic reticulum.
J, Biol. Chem. 265:13472-13483.
Oyamada, H., T. Murayama, T. Takagi, M. Iino, N. lwabe, T. Miyata, Y.
Ogawa, and M. Endo. 1994. Primary structure and distribution of ryanodinebinding protein isoforms of the bullfrog skeletal muscle. J. BioL Chem. 269:
17206-17214.
Percival, A.L., A.J. Williams, J.L. Kenyon, M.M. Grinsell, J.A. Airey, and J.L.
Sutko, 1994. Chicken skeletal muscle ryanodine receptor isoforms: ion channel properties. Biophys. J. 67:1834-1850.
Radermacher, M., V. Rao, R. Grassucci, J. Frank, A.P. Timerman, S. Fleisher,
and T. Wagenknecht. 1994. Cryo-electron microscopy and three-dimensional reconstruction of the calcium release channel/ryanodine receptor
from skeletal muscle. Z Cell Biol. 127:411-423.
Rios, E., J. Ma, and A. Gonzalez. 1991. The mechanical hypothesis of excitation-contraction (EC) coupling in skeletal muscle. 3". Mus. Res. Cell Motil. 12:
127-135.
Rushforth, A., and P. Anderson. 1996. Splicing removes the Caenorhabditis elegans transposon Tcl from most mutant Pre-mRNAs. Mol. Cell. Biol. 16:422429.
Rushforth, A.M., B. Saari, and P. Anderson. 1993. Site-selected insertion of the
transposon Tcl into a Caenorhabditis elegans myosin light chain gene. Mol.
Cell. BioL 13:902-910.
Sakube, Y., H. Ando. and H. Kagawa. 1993. Cloning and mapping of a ryanodine receptor homolog gene of Caenorhabditis elegans. Ann. N Y Acad. ScL
707:540-545.
Somlyo, A.P.. and A.V. Somlyo. 1994. Signal transduction and regulation in
smooth muscle, Nature (Loud.), 372:231-236.
Takeshima, H., M. Lino, H. Takekura, M. Nishi, J. Kuno, O. Minowa, H. Takano, and T. Noda. 1994a. Excitation-contraction uncoupling and muscular
degeneration in mice lacking functional skeletal muscle ryanodine-receptor
gene. Nature (Lond.). 369:556-559.
Takeshima. H., M. Nishi, N. lwabe, T. Miyata, T. Hosoya, I. Masai, and Y.
Hotta. 1994b. Isolation and characterization of a gene for a ryanodine receptor/calcium release channel in Drosophila melanogaster. FEBS Lett. 337:8147.
Takeshima, H., S. Nishimura, T. Matsumoto, H. lshida, K. Kangawa, N. Minamino. H. Matsuo, M. Ueda, M. Hanaoka, T. Hirose, and S. Numa, 1989.
Primary structure and expression from complementary DNA of skeletal
muscle ryanodine receptor. Nature (Loud.). 339:439-445.
Takeshima, H., T. Yamazawa, T. Ikemoto, H. Takeura, M, Nishi, T. Noda, and
M. Iino. 1995. Ca2÷-induced Ca2+-release in myocytes from dyspedic mice
lacking the type-1 ryanodine receptor. E M B O (Eur. MoL BioL Organ.) J. 14:
2999-3006.
Varro, A., N. Negretti, S.B. Hester, and D.A. Eisner. 1993. An estimate of the
calcium content of the sareoplasmic reticulum in rat ventricular myocytes.
Pfliigers Arch. 423:158-160.
Walrond, J.P,. I.S. Kass, A.O.W, Stretton, and J.E, Donmoyer. 1985. Identification of excitatory and inhibitory motorneurons in the nematode Ascaris by
electrophysiological techniques, Z Neurosci. 5:l-8.
Waterston. R., C. Martin, M. Craxton, C. Huynh, C. Coulson, L, Hillier, R.
Durbim P. Green, R. Shownkeen, N. Halloran et aL 1992. A survey of expressed genes in Caenorhabditis elegans. Nature Genet. 1:114-123.
Weisblat, D.A., L. Byerly, and R.L. Russell. 1976. Ionic mechanisms of electrical activity in somatic muscle of the nematode Ascaris lumbricoides, J.
Comp. Physiol. 111:93-113.
Williams, B,D., B. Schrank, C, Huynh, R. Shownkeen. and R.H. Waterston.
1992. A genetic mapping system in Caenorhabditis elegans based on polymorphic sequence-tagged sites. Genetics. 131:609-624.
Wood, W.B. 1988. The nematode Caenhorhabditis elegans. Cold Spring Harbor
Laboratory, Cold Spring Harbor, NY. pp. 1-16, 281-336.
Yano, M., R. El-Hayek, and N. lkemoto. 1995. Conformational changes in the
junctional foot protein/Ca 2+ release channel mediate depolarization-induced
Ca 2+ release from sarcoplasmic reticulum. Z Biol. Chem. 270:3017-302l.
Zwaal, R.R,. A. Broeks, J. van Meurs, J. Groenen, and R.H.A. PLasterk. 1993.
Target-selected gene inactivation in Caenorhabditis elegans by using a frozen
transposon insertion mutant bank. Proc. NatL Acad. Sci, USA. 90:7431-7435.
W e t h a n k Hiro Kagawa and Yasuji Sakube for their generous collaboration over the course of this work. Alice Rushforth's advice was instrumental in the isolation of insertion and deletion mutations. We thank Jeff
Morrisette and D o n n a Vaughn for assistance with binding assays. Some of
the strains used in this work were provided by Caenorhabditis Genetics
Center.
E.B. Maryon was supported by an A m e r i c a n Cancer Society postdoctoral fellowship. This work was supported by National Institutes of Health
research grant GM30132 to P. Anderson.